Today’s article was going to be a breakdown of facts and figures stemming from an article published in the Nation on Thursday entitled “Kenya tops global graft ranking”. “Great” we thought, Kenya is the best place to be for some things such as, err… Theft, fraud, corruption, money laundering…

However after reading the article and passing it around the office, something didn’t quite add up. So we re-read it, then went to the sources of the article, two publications from PricewaterhouseCoopers (PwC) one, a large well based global economic study and the second which is a specific breakdown of the first study but focusing on Kenya (more about these later).

Now lets take a diversion and consider a small book written in the 1950’s but still an essential primer for students, “How to Lie with Statistics” by Darrell Huff. In it is introduction it has this anecdotal story (and remember this was the 1950’s):

“With prospects of an end to the hallowed old British measures of inches and feet and pounds, the Gallup poll people wondered how well known its metric alternative might be. They asked in the usual way, and learned that even among men and women who had been to a university, 33 per cent had never heard of the metric system.

Then a Sunday newspaper conducted a poll of its own and announced that 98% of its readers knew about the metric system. This, the newspaper boasted, showed ‘how much more knowledgeable’ its readers were than people generally.

How can two polls differ so remarkably?

Gallup interviewers had chosen, and talked to a carefully selected cross-section of the public. The newspaper had naively, and economically, relied upon coupons clipped, filled in and mailed in by its readers. Its not hard to guess that most of those readers who were unaware of the metric system had little interest in it or the coupon; and they selected themselves out of the poll by not bothering to clip or participate. This self-selection produced in statistical terms, a biased or unrepresentative sample of just the sort that has led, over the years, to an enormous number of misleading conclusions”

Bearing this story in mind lets get back to the PWC reports.

There are two documents which can be downloaded, “Cybercrime: protecting against the growing threat. Global Economic Crime Survey” and “A step ahead: Economic Crime in Kenya” Both documents come from a huge worldwide survey carried out by PWC and both focus on business economic crime and cybercrime (more on cybercrime next Sunday). The documents are full of sound business advice, and point to key areas of concern, where businesses globally and in Kenya should strengthen their defences, particularly in the areas of cybercrime.

However there are two areas of concern to the Forum team, the reasons we felt uneasy about the original Nation article.

Firstly there is the sampling, 3877 businesses across 78 countries is a massive amount of work, 91 in Kenya alone. But who in these businesses where they asking? Well 52% were described as senior executives, whilst the other 48% were described as non-senior executives, such as the head of business unit, head of department, manager or senior vice president. Forgive us for sounding naive, but that means everyone they polled were major players within the organization. Of these organizations 67% had over 200 employees, so not small or poor companies.

Secondly, is the self selection problem alluded to in the story above. “Amongst respondents in Kenya who reported incidences of economic crime 73% named asset misappropriation as the most prevalent form..” And this is where the big figure (73%) published in the Nation as “Incidences of economic crimes” comes from …which puts Kenya at the top of the list.

It may be true, but if I were a manager running a tidy honest business with no theft, I wouldn’t have bothered to reply to this survey; I’d have nothing to say… this figure could well be misleading in so far of what we are not being told. Elsewhere in the global survey, the charts are qualified by “% respondents” so bear this in mind when reading. The survey can only tell you what the respondents say, not what the non-responders say.

And here’s the bit that has been bugging us from the start, and refers back to the first point. It is from “A Step Ahead” under the heading

“Who is committing the fraud and how do organisations deal with the perpetrators?”

Most fraud is committed by internal fraudsters, according to 68% of Kenya respondents and compared to 59% in Africa and 56% globally. This follows the trend in 2009 where 70% of Kenyan respondents identified perpetrators as internal, with 68% in Africa and 53% globally stating the same. The most common perpetrators of fraud in Kenya are junior staff members (42%) and middle management (42%) which is a trend that is replicated globally.

So there you have it. If you poll 100% of senior executive staff and management, they’ll tell you that almost anyone but themselves steals the loot.

And that dear readers is an example of ‘sampling bias’.